CINXE.COM
PyPy Speed
<!DOCTYPE html> <html> <head> <title> PyPy Speed </title> <meta name="description" content="A performance analysis tool for software projects. It shows performance regresions and allows comparing different applications or implementations"> <meta name="keywords" content="performance, test, plots, charts"> <meta charset="UTF-8"> <link href="https://speed.pypy.org/static/css/main.css" rel="stylesheet" type="text/css"> </head> <body> <div id="container"> <div id="title" class="clearfix"> <a href="/"> <img src="https://speed.pypy.org/static/images/logo.png" height="48" alt="logo"/> </a> <h2>SPEED CENTER</h2> <ul id="links" class="inline"> <li><a href="/">Home</a></li><li><a href="/about/">About</a></li> </ul> </div> <div id="wrapper"> <div id="nav"> </div> <div id="workarea" class="clearfix"> <div id="presentation" class="clearfix"> <a href="/changes/"> <div id="changes" class="menubox"> <h2>Changes</h2> <p>Track performance changes in the latest revisions</p> </div> </a> <a href="/timeline/"> <div id="timeline" class="menubox"> <h2>Timeline</h2> <p>Analyze performance over time</p> </div> </a> <a href="/comparison/"> <div id="comparison" class="menubox"> <h2>Comparison</h2> <p>Compare different executables and revisions</p> </div> </a> <br /> <div id="reports"></div> <div id="historical"> <h3>How fast is PyPy3.11?</h3> <div id="baseline-comparison-plot"></div> <p class="plot-caption">Plot 1: The above plot represents PyPy3.11 (pypy3.11-jit-64) benchmark times normalized to cpython. Smaller is better.</p> <p>It depends greatly on the type of task being performed. The geometric average of all benchmarks is <span id="geomean"></span> or <strong id="geofaster"></strong> times <em>faster</em> than cpython</p> <h3>How has PyPy performance evolved over time?</h3> <div id="historical-plot"></div> <p class="plot-caption">Plot 2: Speedup compared to cpython, using the inverse of the geometric average of normalized times, out of <span id="num_of_benchs"></span> benchmarks (see <a href="http://dl.acm.org/citation.cfm?id=5673" title="How not to lie with statistics: the correct way to summarize benchmark results">paper</a> on why the geometric mean is better for normalized results).</p> </div> </div> </div> </div> <div class="footer">Powered by <a href="https://github.com/python/codespeed/">Codespeed</a>, <a href="http://www.djangoproject.com/">Django</a> and <a href="http://www.python.org">Python</a></div> </div> <script type="text/javascript" src="https://speed.pypy.org/static/js/jquery-1.12.3.min.js"></script> <script type="text/javascript" src="https://speed.pypy.org/static/js/codespeed.js"></script> <link rel="stylesheet" type="text/css" href="https://speed.pypy.org/static/js/jqplot/jquery.jqplot.min.css" /> <script type="text/javascript" src="https://speed.pypy.org/static/js/jqplot/jquery.jqplot.min.js"></script> <script type="text/javascript" src="https://speed.pypy.org/static/js/jqplot/jqplot.barRenderer.min.js"></script> <script type="text/javascript" src="https://speed.pypy.org/static/js/jqplot/jqplot.canvasAxisTickRenderer.min.js"></script> <script type="text/javascript" src="https://speed.pypy.org/static/js/jqplot/jqplot.categoryAxisRenderer.min.js"></script> <script type="text/javascript" src="https://speed.pypy.org/static/js/jqplot/jqplot.canvasAxisLabelRenderer.min.js"></script> <script type="text/javascript" src="https://speed.pypy.org/static/js/jqplot/jqplot.pointLabels.min.js"></script> <script type="text/javascript" src="https://speed.pypy.org/static/js/jqplot/jqplot.canvasTextRenderer.min.js"></script> <script type="text/javascript"> function updateReportTables() { //Add permalink events to table rows $("div#reports table tbody tr").each(function() { $(this).click(function () { window.location = $(this).find("td:eq(0) label").text(); }); }); //Add hover effect to rows $("div#reports table tbody tr td").hover(function() { $(this).parents('tr').addClass('highlight'); }, function() { $(this).parents('tr').removeClass('highlight'); }); } function numOrd(a, b){ return (a-b); } function renderplot(data) { var plotoptions = {}; var plotdata = [new Array(), new Array()]; var labels = new Array(); var benchmarks = new Array(); if (data === null || data.length === 0) { $("#baseline-comparison-plot").html(getLoadText('Error retrieving data', 0)); return 1; } if (typeof data === 'string' || data instanceof String) { $("#baseline-comparison-plot").html(getLoadText('Error retrieving data ' + data, 0)); return 1; } var trunk_geomean = 1; var tagged_data = new Array(); for (i in data['tagged_revs']) { tagged_data[i] = new Array(); } for (var bench in data['benchmarks']) { var relative_value = 0; benchname = data['benchmarks'][bench]; var add_to_tagged_data = true; for (var i in data['tagged_revs']) { var rev = data['tagged_revs'][i]; if (data['results'][benchname][rev] === 0) { add_to_tagged_data = false; } if (add_to_tagged_data === false) { break; } relative_value = data['results'][benchname][rev]/data['results'][benchname][data['baseline']]; tagged_data[i].push(relative_value) } // Only add benchmark if there are no 0 values if (add_to_tagged_data === false) { continue; } // First add benchmark benchmarks.push(benchname); // Add latest results and baseline relative_value = data['results'][benchname]['latest']/data['results'][benchname][data['baseline']]; plotdata[0].push(relative_value); plotdata[1].push(1.0); labels.push(relative_value.toFixed(2)); if (relative_value > 0 && !isNaN(relative_value)) { trunk_geomean *= relative_value; } } trunk_geomean = Math.pow(trunk_geomean, 1/plotdata[0].length); var geofaster = 1/trunk_geomean; $('#geomean').html(trunk_geomean.toFixed(2)); $('#geofaster').html(geofaster.toFixed(1)); // Render first plot plotoptions1 = { legend:{show:true}, seriesDefaults: { showMarker: false, rendererOptions:{barPadding: 2, barMargin:5} }, axesDefaults: { tickRenderer: $.jqplot.CanvasAxisTickRenderer }, series:[ { label: 'latest', renderer:$.jqplot.BarRenderer, pointLabels:{labels: labels} }, { label: data['baseline'], pointLabels:{show: false} } ], axes: { xaxis: { renderer: $.jqplot.CategoryAxisRenderer, ticks: benchmarks, tickOptions: {angle: -70} }, yaxis:{ ticks: [0, 0.5, 1, 1.5], tickOptions:{formatString: '%.2f'} } } }; plot1 = $.jqplot("baseline-comparison-plot", plotdata, plotoptions1); // Prepare and render second plot var geomeans = [1.0]; var num_of_benchs = 0; for (var i in data['tagged_revs']) { num_of_benchs = tagged_data[i].length; var tempgeo = 1; for (var j in tagged_data[i]) { value = tagged_data[i][j]; if (value > 0 && !isNaN(value)) { tempgeo *= value; } } tempgeo = Math.pow(tempgeo, 1/tagged_data[i].length); tempgeo = 1/tempgeo; geomeans.push(tempgeo); } geomeans.push(1/trunk_geomean); var ticks = [data['baseline']]; for (var i in data['tagged_revs']) { ticks.push(data['tagged_revs'][i]); } ticks.push('latest PyPy3.11'); var geolabels = new Array(); for (var i in geomeans) { geolabels.push(geomeans[i].toFixed(2) + "x"); } $('#num_of_benchs').html(num_of_benchs) plotoptions2 = { seriesDefaults: { renderer:$.jqplot.BarRenderer, showMarker: false }, axesDefaults: { tickRenderer: $.jqplot.CanvasAxisTickRenderer }, series:[ { pointLabels:{labels:geolabels} } ], axes: { xaxis: { renderer: $.jqplot.CategoryAxisRenderer, ticks: ticks, tickOptions: {angle: -40} }, yaxis:{ min: 0, tickOptions:{formatString:'%.1f'} } } }; plot2 = $.jqplot("historical-plot", [geomeans], plotoptions2); } $(function() { $.getJSON("/historical/json/", renderplot); }); </script> </body> </html>